PROJECT SUMMARY Schizophrenia (SZ) is a devastating psychiatric illness with a complex etiology. While genetic sequence variation clearly contributes to SZ development, knowledge of DNA methylation patterns may provide complementary information. First, methylation studies may provide additional insights into disease processes. For example, as methylation may directly regulate gene expression and is mutable in postmitotic tissues, it provides functional information unobtainable by sequence alone and can account for clinical phenomena such as the dynamic course of illness including cycles of remissions and relapses. Second, the translational potential of methylation studies is profound. For example, pathogenic environmental events may leave methylation signatures in blood, traces of which can be preserved during cell division. As methylation marks can be measured cost-effectively in (histone-free) genomic DNA, they can potentially increase the predictive power of algorithms based on sequence variation. Two complications hamper inferences in methylation studies of SZ. First, there are many possible differences between SZ cases and controls that may affect the methylome including lifestyle differences, disease induced psychological stress, smoking and antipsychotic use. Rather than methylation affecting SZ susceptibility, the direction of effect could be reversed with the disease causing methylation changes. Second, the procurement of brain tissue is not possible in living patients. It is therefore important to study the methylome using a combination of blood samples, which can reveal useful biomarkers, and in brain tissue, which can help identify biologically relevant functions regulated by methylation. To address these complications we propose to study the methylomic profiles of SZ cases/controls in blood samples collected at birth. Because these samples were collected years prior to any SZ symptoms, it is logically impossible for SZ disease related confounders to cause the associations. Next, to generate causal hypotheses that can potentially be followed up with functional experiments, we integrate the pre SZ onset findings with methylation and transcription data from post SZ onset blood and brain samples using a logically rigorous analytical framework. As detailed knowledge is lacking about disease relevant methylation sites, we propose to assay all ~27 million CpGs in the human genome in all our samples. Successful completion of this proposal will identify biomarkers and functional methylation marks for SZ generally, and may specifically identify markers with potential to predict SZ risk. Such findings would be of considerable value for improving SZ treatment and care.